Image retrieval is typically achieved by tagging the images with text and then using text based document retrieval methods to find images relevant to query words entered by a user. Other approaches, based on analyzing the content of images automatically are more complex.
As the availability of image files on the internet and from other sources is continually increasing there is an ongoing need for efficient and effective image retrieval systems.
The embodiments described below are not limited to implementations which solve any or all of the disadvantages of known image retrieval systems.